US 5,625,124 AGrant
Animal Model for Helicobacter Pylori Infection
Issue Date:1997-04-29
•9 Claims
•6 Drawing Sheets
Abstract
Transgenic non-human animals are described which serve as a model for H. pylori infection of epithelial cells of the stomach and small intestine. The gut epithelial cells of the transgenic animals express one or more surface carbohydrate antigens which act as receptors for the bacterium H. pylori, a known causative agent of acid peptic disease, such as gastritis, stomach ulcers, duodenal ulcers, and strongly correlated with the development of gastric neoplasia. Methods for making and using the transgenic animals are also disclosed. The transgenic animals can be used to screen for compounds and conditions which block binding of H. pylori to the gut epithelium and/or ameliorate the H. pylori-associated pathogenesis of acid peptic disease and gastric adenocarcinoma.
Metadata
Assignee
- Washington University
Inventors
- Per Falk
- Jeffrey I. Gordon
Application Information
Application Number:US 2734117
Filing Date:1994-07-11
Priority Date:1994-07-11
Art Unit:184
Classifications
IPC:
C12N 500A61K 4900G01N 33567
Field of Search:
4244358009359;9.27.21;172.1;172.3;240.22;DIG. 159;60;70
Patent Drawings (6 sheets)
Description
This application is generally in the field of transgenic animals, and in particular, transgenic animals are described which can serve as an animal model for Helicobacter pylori infection.
Background of the Invention
Recently, it has been found that the bacterium Helicobacter pylori is a causative agent of human chronic gastritis, gastric and duodenal ulcers, as well as gastric adenocarcinoma in humans. (Blaser, Trends Microbiol., 7: 255-260 (1993)). H. pylori produces sialic acid-specific agglutinins enabling the bacterium to bind sialyl residue containing carbohydrates. However, studies using an in vitro adherence assay and other data have revealed that strains of H. pylori isolated from patients with acid peptic disease bind preferentially to the surface mucous and pit cells in the human gastric epithelium. This binding does not depend on the presence of sialic acid epitopes, but fucose residues constitute an essential recognition element of the epithelial cell's adhesion receptor (Falk et al., Proc. Natl. Acad. Sci. USA, 90: 2035-2039 (1993)). Additional studies have indicated that adherence of clinical isolates of H. pylori (P466 and WV229) to human gastric surface mucous cells occurs via the Le.sup.b antigen (Boren, et al., Science, 262: 1892-1895 (1993)), a fucosylated carbohydrate expressed by members of the pit cell lineage (Sakamoto, et al., Cancer Res., 49: 745-752 (1989); Falk, et al., Proc. Natl. Acad. Sci. USA, 90: 2035-2039 (1993). Solid phase binding and inhibition assays also indicated that the human H-1 blood group antigen, a monofucosylated carbohydrate that biochemically defines the blood group O phenotype, may also serve as a specific receptor for bacterial binding. In addition, such binding appears to be specific for the human gut, because binding to comparable cell lineages in the mouse, rat and dog is either significantly weaker, as in Sprague-Dawley rats, or not detectable, as in FVB/N mice. Thus, H. pylori has been found to preferentially bind to distinct molecules, such as the Le.sup.b and H-1 antigens, and to sialic acid residues found on glycoproteins on the surface of gastric epithelial cells (Boren et al., Science, 262: 1892-1895 (1993); Evans et al., J. Bacteriol., 175: 674-683 (1993)).
Combined chemotherapy with antibiotics (tetracycline and metronidazole), H.sub.2 blockers (ranitidine), and bismuth subsalicylate have been used with some success for treating H. pylori-associated ulcers, although relapse remains a clinical problem (Graham, et al., Ann. Int. Med., 116: 705-708 (1991)). The exact mechanism by which this bacterium causes chronic active gastritis, gastric and duodenal ulcers and gastric adenocarcinoma remains unclear. It is well documented that H. pylori infection of gut epithelial is strongly correlated with development of gastric adenocarcinoma (six to twelve times increased risk according to Forman, et al., Lancet, 341: 1359-1363 (1993)). Gastric adenocarcinoma is the fourth leading cause of cancer death worldwide (Levine and Price, In Helicobacter pylori infection (Northfield, Mendall, and Goggin eds.), pp. 88-89 (Kluwer Acad. Publ., Boston (1993)).
An effective alternative to antibiotic therapy that is curative and/or prevents H. pylori-associated disease has not been developed. An animal model for H. pylori infection of the stomach and/or intestinal tract would thus be highly desirable not only for detailed in vivo studies of H. pylori-associated diseases caused by wild type and genetically engineered strains, but also for the systematic development and screening for new antibiotics, non-antibiotic compounds, and various other therapies employing natural products and chemically synthesized compositions of matter which are effective in blocking H. pylori adhesion, infection, or the subsequent development of H. pylori-associated disease states. In addition, an animal model for H. pylori infection in the stomach and intestinal tract could be used in developing and assessing effective gene therapies against H. pylori-associated gastritis, ulcers, carcinomas, and presumably other H. pylori-associated diseases, as well as examining the efficacy of various methods used to immunize the host so as to prevent or ameliorate H. pylori infection.
It is therefore an object of the present invention to provide an animal model for H. pylori infection and pathogenesis.
It is a further object of the present invention to provide a method for screening and evaluating compounds which block H. pylori adhesion to gut endothelium or ameliorate the effects of H. pylori binding on the pathogenesis of acid peptic disease.
Summary of the Invention
Transgenic non-human animals which stably express nucleic acid molecules encoding the human GDP-L-fucose: .beta.-D-galactoside 2-.alpha.-L-fucosyltransferase (EC 2.4.1.69, also referred to as .alpha.1,2 fucosyl transferase or .beta.1,2FT) and/or the human GDP-L-fucose: .beta.-D-N-acetylglucosaminide 3/4-.alpha.-L-fucosyltransferase (also called .alpha.1,3/4 fucosyltransferase or .alpha.1,3/4 FT) preferentially exhibit fucosylated epitopes, such as the H-1 and Le.sup.b antigens and related structures, on the surfaces of gut epithelial cells. These animals can serve as an animal model for H. pylori adhesion and associated diseases and be used to systematically screen for compounds that block H. pylori adhesion, or in other ways to intervene in H. pylori infection of the stomach or intestinal tract.
As demonstrated by the example, transgenic mice have been made by constructing a transgene including a gut epithelial cell specific promoter and a nucleotide sequence encoding a human fucosyltransferase (FT).
Brief Description of the Drawings
FIG. 1 is a diagram illustrating the formation of the H type 1 blood antigen and the Le.sup.a and Le.sup.b antigens mediated by .alpha.1,2 FT and .alpha.1,3/4 FT.
FIG. 2 is a diagram of the recombinant pBLUESCRIPT.TM. II (Stratagene, La Jolla, Calif.) SK-.alpha.1.2 FT vector showing relative positions of selected restriction endonuclease sites.
FIG. 3 is a diagram of the recombinant pBLUESCRIPT.TM. II (Stratagene, La Jolla, Calif.) SK-.alpha.1.3/4 FT vector showing relative positions of selected restriction endonuclease sites.
FIG. 4 is a diagram of pLPN.DELTA.B2 showing relative position of nucleotides (nt)-596 to +21 of the rat Fabpl gene which corresponds to nucleotides 1 to 617 of Sequence ID No. 5, linked to nucleotide +3 to +2150 of the human growth hormone (hGH) gene which corresponds to nucleotides 620 to 2771 of Sequence ID No. 5, and pgK-neo. The pgK-neo marker is a neomycin resistance selection cassette under the control of the phosphoglycerate kinase promoter, but is not a part of the 4 kilobase (kb) gene construct that was injected into embryos described in Example 3. The relative positions of selected restriction endonuclease sites are also shown.
FIG. 5 is a diagram of plasmid LF-.alpha.1.2 FT showing relative positions of nucleotides -596 to +21 of the rat Fabpl gene (nucleotide 1 to 617 of Sequence ID No. 5), nucleotides +3 to +2150 of the hGH gene (nucleotides 620 to 2771 of Sequence ID No. 5), and the 1113 nucleotide .alpha.1,2 FT sequence including the 1096 nucleotide .alpha.1,2 FT coding region, (Larsen, et al., Proc. Natl. Acad. Sci. USA, 87: 6674-6678 (1990)) (nucleotides 104 to 1201 of Sequence ID No. 2), the 5' (upstream of the transcriptional start site) HindIII site and Kozak concensus sequence (AGCTTGCCACC, Sequence ID No. 11) not in the published sequence, and the 3' (downstream of the translational stop codon) NotI (GCGGCCGC) site. The relative positions of selected restriction endonuclease sites are also shown.
FIG. 6 is a diagram of plasmid LF-.alpha.1.3/4 FT showing relative positions of nucleotides -596 to +21 (nucleotides 1 to 617 of Sequence ID No. 5) of the rat Fabpl gene, nucleotides +3 to +2150 (nucleotides 620 to 2771 of Sequence ID No. 5) of the hGH gene, and the 1103 nucleotide .alpha.1,3/4 FT sequence including the 1086 nucleotide coding region (nucleotides 73 to 1158 of Sequence ID No. 4), the 5' (upstream of the start site of transcription) HindIII site and Kozak consensus sequence and the 3' NotI site (analogous to FIG. 5).
Detailed Description of the Invention
H. pylori produces adhesins that enable the bacterium to bind human H antigen or Lewis antigens on lacto-series type 1 chains found on the surface of gut epithelial cells. For an animal to serve as an accurate and reliable animal model of H. pylori gut infection, it must be capable of producing on the surfaces of gut epithelial cells these antigens which function as receptors for H. pylori and, thereby, enable the bacterium to establish an initial attachment and subsequent infection in a manner similar to that in humans. Transgenic animals are described that express the fucosyltransferases (FTs) which enable the gut epithelial cells of the animals to produce one or more of the known H. pylori receptor antigens, such as the H type 1 and Le.sup.b antigens, on their cell surfaces. These transgenic animals can therefore serve as an animal model for H. pylori infection of the stomach and/or intestinal tract. This animal model is particularly useful to systematically test or screen for compounds which block H. pylori adhesion and infection and which are useful in preventing or treating H. pylori-associated disease states such as chronic gastritis, ulcers, and gastric adenocarcinoma.
Synthesis of Complex Carbohydrate Receptors
Complex carbohydrates are ubiquitous in the plasma membranes of eukaryotic cells. All complex carbohydrates are composed of highly variable sequences of monosaccharide units (primarily glucose, galactose, N-acetylglucosamine, N-acetylgalactosamine, fucose, sialic acid, and mannose) linked together at different binding positions with different anomerity (.alpha. or .beta.) of the glycosidic linkages. Complex carbohydrates can be anchored to a protein via linkage between the innermost N-acetylglucosamine residue of the carbohydrate and asparagine residues in the peptide core (N-linked, the most common in membrane-associated glycoproteins), or via linkage between the innermost N-acetylgalactosamine residue of the carbohydrate to serine or threonine in the peptide core (O-linked or mucin-type). Glycosphingolipids, the glycosides of N-acylsphingosine, are the lipid-bound glycoconjugates of eukaryotic cell membranes. They carry oligosaccharides similar to their glycoprotein counterparts.
The synthesis of complex carbohydrates is mediated by glycosyltransferases, a group of highly specific enzymes (generally associated with the Golgi apparatus) that add activated (nucleotide conjugated) sugars to carbohydrate chains in a step-by-step elongation process. These enzymes are highly specific with respect to both the particular sugar "substrate" to be added and the acceptor molecule to which the substrate monosaccharide is transferred. The expression of glycosyltransferases, and hence, the composition of cell surface carbohydrates, is regulated in a species-specific, cell lineage-specific, and developmental stage-specific manner (Paulson and Colley, J. Biol. Chem., 264: 17615-17618 (1989)).
Cell surface carbohydrates have been shown to play a key role in cell-cell recognition events mediated by protein-carbohydrate and carbohydrate-carbohydrate interactions (for example, Sharon and Lis, Science, 246: 227-234 (1989)). Cell surface carbohydrates can also function as receptor molecules for a large number of bacteria, bacterial toxins, viruses, and protozoa (see, Sharon and Lis, Science, 246: 227-234 (1989); Karlsson, Ann. Rev. Biochem., 58: 309-350 (1989)).
Fucose residues play a key role in the histoblood group antigenic phenotype of eukaryotic cells, because they are a crucial component of the epitopes that define the A, B, O (H), and Lewis blood group antigen systems (see, for example, Clausen and Hakomori, Vox Sang., 56: 1-20 (1989)). Fucosylated complex carbohydrates have also been identified as onco-fetal antigens and as adhesion molecules involved, for instance, in the early stages of an inflammatory response and in host-microbial interactions. In addition to H. pylori, Candida albicans (Tosh and Douglas, Infect. Immun. 60: 4734-4739 (1992)) and Vibrio cholerae (Sanchez and Jonson, Acta Pathol. Microbiol. Immunol. Scand., 98: 353-357 (1990)) have been shown to express fucose-specific adhesion molecules. The syntheses of fucosylated sugars are regulated by the expression of specific fucosyltransferases (FTs). The genes coding for six different human FTs have been described. The first FT genes to be cloned were the human Lewis gene (Le) encoding human .alpha.1,3/4 FT (see, Kukowska-Latallo, et al., Genes Dev., 4: 1288-1303 (1988)) and the human H gene encoding .alpha.1,2 FT (Larsen et al., Proc. Natl. Acad. Sci. USA, 87: 6674-6678 (1990)). In humans, the .alpha.1,3/4 FT can generate the sequence Gal.beta.1-3/4(Fuc.alpha.1-3/4)GlcNAc (FIG. 1) in complex carbohydrates on cellular membrane surfaces and is responsible for producing the stage-specific embryonic antigen 1 (SSEA-1) and the Lewis blood group antigens (Kukowska-Latallo et al., Genes Dev., 4: 1288-1303 (1990)). The .alpha.1,2 FT can add an L-fucose residue in an .alpha.1,2 position to a terminal galactose to generate the sequence Fuc.alpha.1-2Gal.beta.- in such surface carbohydrates and is responsible for generating the blood group H antigen, which is the molecule that structurally defines the blood group O phenotype (Clausen and Hakomori, Vox Sang., 56: 1-20 (1989)).
The .alpha.1,2 FT cDNA sequence predicts a 365 amino acid protein product with the characteristics of a type II transmembrane glycoprotein (see, Larsen et al., Proc. Natl. Acad. Sci. USA, 87: 6674-6678 (1990)). The .alpha.1,3/4 FT is a 361 amino acid, type II transmembrane glycoprotein that was originally isolated using an expression cloning procedure based on the ability to produce SSEA-1 on the surface of transfected COS-1 cells (see, Kukowska-Latallo et al., Genes Dev., 4: 1288-1303 (1990)). SSEA-1 is an oligosaccharide expressed during the murine preimplantation period and is thought to be involved in the process of compaction (Bird and Kimber, Dev. Biol., 104: 449-460 (1984)).
DNA Sequences Encoding .alpha.1,2 FT and .alpha.1,3/4 FT
The cDNA sequences encoding the .alpha.1,2 FT and the .alpha.1,3/4 FT have previously been cloned and their sequences published (Larsen et al., Proc. Natl. Acad. Sci. USA, 87: 6674-6678 (1990) (.alpha.1,2 FT); Kukowska-Latallo et al., Genes Dev., 4: 1288-1303 (1990) (.alpha.1,3/4 FT)). These are shown as Sequence ID No. 2 and 4, respectively.
Selection of Animals Useful for Construction of Transgenic Animals, and the Resulting Transgenic Animals Expressing the H1 and/or Lewis Antigen
Animals that would be particularly useful as animal models are those which are easily bred, raised and readily manipulated genetically and biochemically. Such animals include the mouse, rat, gerbil, rabbit, hamster and guinea pig.
The following illustrates the criteria useful for selecting an animal species and strain used for creating an H. pylori infection model. General criteria include expression of potential acceptor molecules that could be used by human FTs to generate the H1 and/or Lewis antigen in the animal to be used; followed by expression of the H1 and/or Lewis antigen in the transgenic animal.
For example, the FVB/N strain of mice was chosen to generate this model as a result of several observations. First, the pit and enterocytic cell lineages of FVB/N mice contain potential acceptor molecules that could be used by human FTs to generate the H1 and/or Lewis antigens. Formation of the H antigen requires a carbohydrate core chain acceptor with a terminal galactose, preferably members of the lacto-series, and a soluble GDP-L-fucose substrate that is accessible to .alpha.1,2 FT for the transglycosylation reaction (see FIG. 1 and Larsen, et al., Proc. Natl. Acad. Sci. USA, 87: 6674-6678 (1990)). An in situ binding assay which uses fluorescent labeled galactose-specific lectins revealed the presence of such acceptors in the surface mucous, pit and enterocytic lineages of adult FVB/N mice (Falk, et al., Am. J. Physiol. (Gastroenterolo Liver Physiol., 266), G987-G1003, (1994)). For example, the lectin Trichosantes kirilowii (TKA) which binds to carbohydrate epitopes containing a terminal .beta.-linked galactose, reacts with these FVB/N cell lineages in the stomach and intestine. Formation of the Lewis antigens requires a carbohydrate acceptor containing terminal Gal.beta.1,3/4-GlcNAc or Fuc.alpha.1,2 Gal.beta.1,3/4-GlcNAc sequences with the GlcNAc serving as the site of attachment of L-fucose residues through the action of .alpha.1,3/4 FT (see FIG. 1 and Kukowska-Latallo, et al., Genes Develop., 4: 1288-1303 (1990)). The in situ binding assay indicated that the surface mucous, pit and enterocytic lineages contain such acceptor molecules. Further evidence was shown by binding of these cells by both of the lectins Erythrina christagalli (ECA) and Datura stramonium (DSA), which recognize Gal.beta.4GlcNAc.beta.- containing epitopes. These observations indicated that it should be possible to produce H1 and/or Lewis epitopes by expressing either one or both of the aforementioned human FTs in these FVB/N mouse gut epithelial cells.
The H1 and/or Lewis antigens can be readily detected in transgenic mice of this strain using lectins and the in situ binding assay. A normal adult FVB/N mouse that has not been genetically engineered to produce the human FTs, does not produce detectable levels of fucosylated glycoconjugates in their small intestinal enterocytes, a cell lineage that supports expression of reporter genes, such as fucosyltransferase with the Fabpl.sup.-596 to +21 promoter, as described below. Novel fucosylated carbohydrate epitopes could therefore be detected in these cells using fucose-specific lectins and monoclonal antibodies, as described below. In the FVB/N strain of mice, the fucose-specific lectins, Ulex europaeus agglutinin I (UEA-1), Anguilla anguilla agglutinin (AAA), and the Lotus tetragonolobus agglutinin only stain gastric surface mucous/pit cells and intestinal Paneth cells, M-cells, and subsets of goblet and enteroendocrine cells, but not enterocytes (Falk, et al., Am. J. Physiol. (Gastroenterol. Liver Physiol. 29, 266, G987-G1003), (1994)). Moreover, these antigens could be detected in transgenic FVB/N mice with other reagents which recognize fucose-containing glycoconjugates, such as commercially available monoclonal antibodies (MAbs) specific for blood group ABO(H) and Lewis antigens, or with FITC-labeled H. pylori and fixed sections of stomach, duodenum, jejunum, ileum and colon (Falk, et al., Proc. Natl. Acad. Sci. USA, 90: 2035-2039 (1993)).
Design of .alpha.1,2 FT and .alpha.1,3/4 FT Transgenes
a. Tissue-specific Promoters
To increase the accuracy and reliability of an animal model for H. pylori adhesion and infection, it is desirable that proper genetic controls direct expression of cell surface receptors for bacterial adhesins in the relevant cell lineages of the gut epithelium, such as in surface mucous and pit cells. A preferred example of a suitable promoter is the promoter sequence Fabpl.sup.-596 to +21 contained within nucleotides (nt) -596 to +21 of the rat liver fatty acid binding protein gene (Fabpl) (nucleotides 1 to 617 of Sequence ID No. 5) which can direct foreign gene expression to the pit cell lineage of the mouse gastric epithelium, to proliferating and nonproliferating cells in intestinal crypts, as well as to the four principal differentiated cell lineages distributed along the crypt-to-villus axis of the small intestine (Roth et al., Am. J. Physiol. (Gastrointest. Liver Physiol., 26), 263: G186-G197 (1992); Trahair et al., J. Cell Biol., 109: 3231-3242 (1989); Simon et al., J. Biol. Chem., 268: 18345-18358 (1993), incorporated herein by reference).
The Fabpl promoter sequences or other such tissue-specific promoter sequences are inserted upstream of the coding sequences for .alpha.1,2 FT and/or .alpha.1,3/4 FT to direct synthesis of H and Lewis blood group antigens in the gut epithelium of animals containing the recombinant DNAs as transgenes.
b. Expression Enhancers
Other nucleic acid sequences can also be incorporated into the design of the .alpha.1,2 FT and .alpha.1,3/4 FT transgenes. For example, recombinant DNA molecules can be made in which a cDNA encoding a human FT is inserted downstream of the Fabpl.sup.-596 to +21 promoter sequence and into exon 1 of a human growth hormone gene (hGH, Seeburg, DNA, 1: 239-249 (1982)) sequence consisting of nucleotides +3 to +2150 of the hGH gene (nucleotides 620 to 2771 of Sequence ID No. 5). No human growth hormone should be produced from such recombinant DNAs because the translational initiator Met codon and the first translational stop codon are from the fucosyltransferase DNA, and there is no ribosomal re-entry sequence to re-initiate translation at the downstream initiator Met codon of the hGH gene sequence. The hGH gene sequence insures efficient splicing of the primary transcription product of the transgene, improves the stability of the cytoplasmic fucosyltransferase mRNA (Sandgren, et al., Cell, 61: 1121-1135 (1991)) and allows expression of the transgene to be monitored either by immunocytochemical methods using an antibody directed against the fucosyltransferase or by in situ hybridization using a digoxigenin-labeled hGH oligonucleotide (Simon, et al., J. Biol. Chem., 268: 18345-18358 (1993)).
Production of Transgenic Mice
Transgenic animals can be generated by standard methods in which a genetic construct is introduced by pronuclear injection into a fertilized egg. The most common methods of such direct introduction of gene sequences are microinjection of embryos and injection of embryonic stem cells into blastocysts. The procedures for manipulation of the rodent embryo and for microinjection of DNA are described in detail in Hogan et al., Manipulating the mouse embryo, (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1986), the teachings of which are incorporated herein. Embryonic stem (ES) cells can be manipulated using the method of Gossler, et al., "Transgenesis by means of blastocyst-derived embryonic stem cell lines" Proc. Natl. Acad. Sci. USA, 83: 9065-9069 (1986), the teachings of which are incorporated herein. The use of nucleotides 1 to 617 (Sequence ID No. 5) of rat Fabpl (Fabpl.sup.-596 to +21) in gene expression in embryonic stem cell components of chimeric transgenic mice are described in Hermiston, et al. (Proc. Natl. Acad. Sci. USA, 90: 8871-8875 (1993)) which also describes methods for distinguishing ES cell components of chimeric gut epithelium from normal host blastocyst components by means of their different patterns of glycoconjugate expression.
Typically, when microinjected embryos are to be used to produce transgenic mice, female mice six weeks of age are induced to superovulate, for example, with a 5 IU injection (0.1 cc, intraperitoneally (ip)) of pregnant mare serum gonadotropin (PMSG, Sigma, St. Louis, Mo.), followed 48 hours later by a 5 IU injection (0.1 cc, ip) of human chorionic gonadotropin (hCG, Sigma). Females are placed with males immediately after hCG injection. Twenty-one hours after hCG, the mated females are sacrificed by CO.sub.2 asphyxiation or cervical dislocation and embryos are recovered from excised oviducts and placed in Dulbecco's phosphate buffered saline (DPBS) with 0.5% bovine serum albumin (BSA, Sigma). Surrounding cumulus cells are removed with hyaluronidase (for example, 1 mg/mL). Pronuclear embryos are then washed and placed in a buffer such as Earle's balanced salt solution containing 0.5% BSA (EBSS) in a 37.5.degree. C. incubator with a humidified atmosphere at 5% CO.sub.2, 95% air until the time of microinjection.
Purified linear DNA molecules including an FT transgene construct are purified and diluted to 5 ng/mL in Buffer A (10 mM Tris/0.2 mM EDTA, pH 7.4). The diluted DNA is then filter sterilized and injected into the male pronucleus of newly fertilized mouse oocytes using standard microinjection methods (see, Hogan et al., Manipulating the mouse embryo, pp. 157-173 (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1986)).
Once the embryos are injected, the live embryos are transferred into foster mothers. Foster mothers are produced from randomly cycling adult female mice paired with vasectomized males. Swiss Webster or other comparable strains can be used for this purpose. Recipient females are mated at the same time as donor females. At the time of embryo transfer, the recipient females are anesthetized with an intraperitoneal injection of, for example, 0.015 mL of 2.5% avertin per gram of body weight. The oviducts are exposed by a single midline dorsal incision. An incision is made through the body wall directly over the oviduct. The ovarian bursa is torn, for example, with watchmakers forceps. Embryos to be transferred are placed in DPBS or other comparable buffer and withdrawn into the tip of a transfer pipet (about 10-12 embryos). The pipet tip is inserted into the infundibulum and the embryos transferred. After the transfer, the incision is closed with sutures, and the foster mothers are housed, one or two per cage.
Creating Bi-transgenic Mice That Co-express .alpha.1,2 and .alpha.1,3/4 FTs in Gut Epithelial Cell Lineages
In cells that contain fucosylated glycocojugates (for example, members of the surface mucous and pit lineages of the FVB/N mouse stomach), expression of a single additional human FT could be sufficient for H. pylori . However, in gut epithelial cells lacking endogenous fucosylated glycoconjugates, it is necessary to express both .alpha.1,2 and .alpha.1,3/4 FTs to produce the Lewis b receptor. A relatively easy way to obtain transgenic animals expressing two transgenes is to mate two animals, each of which expresses one of the transgenes (see, for example, Kim et al., J. Cell Biol., 123: 877-893 (1993); Khillan et al., Nucleic Acids Res., 16: 1423-1430 (1988)). For example, a male or female mouse containing a transgene consisting of rat Fabpl.sup.-596 to +21 linked to .alpha.1,2 FT-hGH can be mated to a mouse containing a transgene consisting of rat Fabpl.sup.-596 to +21 linked to .alpha.1,3/4 FT-hGH, to produce progeny containing and expressing both transgenes.
Screening for .alpha.1,2 FT and .alpha.1,3/4 FT Transgene Expression in Transgenic Animals
The expression of .alpha.1,2 FT and .alpha.1,3/4 FT in transgenic animals can be determined by a variety of methods.
1. Screening with nucleic acid probes
Nucleic acid probes can be used to detect the presence of the transgenes in an animal genome. For example, tail samples (0.5-1 cm) can be removed from three week old mice, and the DNA prepared and analyzed by both Southern blotting (using .sup.22 P labeled FT DNA as probes) and PCR (using Fabpl, hGH and/or FT oligonucleotides) to detect transgenic founder (F.sub.0) animals and their progeny (F.sub.l and F.sub.2). In addition, cDNA sequences used to construct the FT transgenes can be used in RNA blot hybridization, RNAse protection, and/or reverse transcriptase PCR assays, to detect FT-hGH mRNA transcripts in the cells of various tissues, especially the cells of gut epithelial tissue.
2. Screening for binding to receptors
Alternatively, new carbohydrate antigen receptors for H. pylori generated on the surface of cells of gut epithelium by the expression of functional .alpha.1,2 FT and/or .alpha.1,3/4 FT can be detected using one or more appropriate lectins or monoclonal antibodies (MAbs), which bind specifically to the H antigen or to a particular Lewis antigen.
a. Lectins
Lectins are proteins which bind specific carbohydrate structures. Such proteins can be used to detect the presence of carbohydrate epitopes in the gut epithelium of a host animal (Falk, et al., Am. J. Physiol. (Gastroenterol. Hepatol., 29), 266, G987-G1003, (1994)). Lectins that are useful for specifically detecting the presence of fucosylated glycoconjugates, such as the H antigen and Lewis antigens, on the surfaces of gut epithelial cells include Ulex europaeus agglutinin I (UEA-I), Anguilla anguilla agglutinin (AAA), and Lotus tetragonolobus agglutinin. (Falk, et al., Am. J. Physiol. (Gastroenterol. Hepatol., 29,) 266, G987-G1003) (1994)).
b. Monoclonal antibodies
Monoclonal antibodies (MAbs) which bind to specific carbohydrate structures, such as H or Lewis antigens are commercially available (for example, DAKO, Carpinteria, Calif.; Immucor, Inc., Norcross, Ga.; Signet Lab. Inc., Dedham, Mass.). As with lectins, MAbs directed against specific carbohydrates can be used to detect the presence of carbohydrate structures on the gut epithelial cells of transgenic and non-transgenic animals. Since most MAbs are produced in mice, they can be directly conjugated with a fluorochrome, such as fluorescein isothiocyanate, tetramethyl-rhodamine isothiocyanate or Cy3.TM., or with digoxigenin groups (see Section c., below) allowing detection with, for example, Fab fragments of sheep anti-digoxigenin (Boehringer-Mannheim, Indianapolis, Ind.) conjugated with fluorescent or enzyme (alkaline phosphatase, horseradish peroxidase) tags.
c. H. pylori or its membrane fragments
As an alternative to using either lectins or MAbs, H. pylori cells can be labelled, for example, with fluorescein isothiocyanate or digoxigenin-3-O-succinyl-.epsilon.-aminocaproic acid N-hydroxysuccinimide ester (DIG-NHS, Boehringer Mannheim, Indianapolis, Ind.), and used as a diagnostic reagent to bind and detect the newly synthesized surface H. pylori receptors on the surface of gut epithelial cells of transgenic animals (see, Falk et al., Proc. Natl. Acad. Sci. USA, 90: 2035-2039 (1993); Boren et al., Science, 262: 1892-1895 (1993)).
Use of .alpha.1,2 FT and .alpha.1,3/4 FT Transgenic Animals
A transgenic animal expressing either or both .alpha.1,2 FT and .alpha.1,3/4 FT generates the H and/or Le.sup.b cell surface carbohydrate antigens and therefore is a powerful model for studying the effects of H. pylori on gut epithelial biology. Such animals constitute the first genetically well-defined and manipulable animal model of H. pylori adhesion and infection. The animal model can be used to systematically screen for compounds that block binding of H. pylori to the gut epithelium and/or ameliorate the effects of H. pylori binding on the pathogenesis of acid peptide disease, for example, gastritis, stomach ulcers, duodenal ulcers, and gut neoplasia, such as stomach adenocarcinoma.
1. Establishment of Infection with H. pylori
H. pylori bacteria can be orally administered to establish an infection in the gut epithelium of transgenic animals which express the H antigen and/or Le.sup.b antigens. The establishment of the H. pylori infection can be confirmed by examining tissue samples for bacteria and/or signs of inflammation, ulceration or carcinoma. Once the bacterial infection has been established, a compound or a series of compounds can then be administered to the infected animal at various times and in various dosages, depending on the particular goals of the screen. In a variation of this procedure, it may be desirable to administer the bacteria with a compound to determine whether, relative to control animals, the compound can effectively prevent in vivo the initial bacterial adhesion and/or the subsequent establishment of infection or pathogenesis.
2. Screening for compounds blocking infection
The efficacy of the compound or compounds can be assessed by examining at selected times the cells of the gut epithelial tissue of the infected animals for the presence or loss of adherent H. pylori bacteria and/or the development, inhibition, or amelioration of ulcer or tumor formation relative to appropriate control mice, for example, untreated H. pylori-infected animals.
This is particularly useful in the testing of vaccines against the bacteria, which are administered to the animal approximately ten days prior to exposure to the bacteria, and thereafter as required to produce immunity against infection or development of a disorder associated with infection.
Since the described transgenic animals express the receptors which bind H. pylori in human gut epithelium, they offer the additional benefit of allowing the efficacy of various drugs or compounds to be readily assessed based on different modes of administration and compound formation. In addition to using the transgenic animals to screen for therapeutic compounds, these animals can also be used to screen for conditions or stimuli which effect a block in or ameliorate H. pylori adhesion, infection and/or associated gut diseases. Such stimuli or conditions include environmental or dietary changes, changing the gastro-intestinal pH, or combinations of various stimuli or conditions which result in stress on the animal or on H. pylori bacteria in the gut. H. pylori can be administered to the described transgenic animals, and the animals then exposed to a selected stimulus or condition, or a combination of stimuli or conditions, to be tested. The gut epithelial tissue of exposed transgenic animals is then examined periodically for a change in the number of adherent H. pylori bacterial and/or the disease state of the epithelial tissue relative to non-exposed control transgenic animals.
Another type of condition that can be tested for its effect on H. pylori adhesion, infection, and/or associated gut disease in the transgenic animals described herein is the induction of an inflammatory response, for example by administering a chemical agent such as dextran sulfate, at various times prior to, during, or after administration of H. pylori to the transgenic animal. The inflammatory agent can be administered orally or by any other mode that results in a gastrointestinal inflammatory response. The severity of the inflammatory response can be controlled by varying the dose and the duration of treatment with the chemical agent.
3. Other Uses of Transgenic Animals Expressing H. pylori Receptor Molecules.
Transgenic animals expressing H. pylori receptor molecules can also be used as a source of epithelial cells which can be cultured and infected in vitro with H. pylori . For example, conditionally immortalized gut epithelial cells can be produced from transgenic mice containing Fabpl.sup.-596 to +21 linked to a gene coding for a temperature sensitive Simian virus 40 T antigen. (TAg) (Jat and Sharp, Mol. Cell Biol., 9, 1672-1681 (1988)). The temperature sensitive TAg is inactive at 37.degree.-40.degree. C. but is active at 33.degree. C. allowing cells that express the protein to continuously proliferate at the latter temperature. Cells carrying the TAg transgene can be harvested from transgenic animals and grown in primary cell culture at 33.degree. C. to establish conditionally immortalized lines. The differentiation programs of cloned, established cell lines can then be induced by switching the incubation temperature to 39.degree. C. If such animals are crossed to animals containing a single FT transgene or to an .alpha.1,2 FT/.alpha.1,3/4 FT bi-transgenic mice creating TAg/FT expressing bi- or tri-transgenic animals that could be a source of H1 and/or Le.sup.b producing gut epithelial culture cell lines. Such cultured cells can be used for rapid in vitro screening protocols for compounds that block H. pylori adhesion and infection. H. pylori adhesion to such cultured cells can readily be determined in vitro, using manual, semi-automated or automated methods as, for example, through the use of commercially available fluorescently labeled antibodies to H. pylori (Harlan Bioproducts for Science, Indianapolis, Ind.). Compounds that are initially identified as inhibitors of H. pylori adhesion or infection in such an in vitro system can then be further screened and studied for their in vivo efficacy against H. pylori using a transgenic animal as described above and below.
Germ-free (see, for example, Granholm, et al., Cytokine, 4, 545-550 (1992)) transgenic animals can be generated by embryo transfer into gnotobiotic females raised in isolator cages. These germ-free transgenic animals can be monocontaminated with laboratory strains of H. pylori or various clinical isolates to directly examine the pathogenesis of H. pylori infection and the efficacy of drugs in modifying the course or extent, of infection and associated gastrointestinal pathology in the absence of a normal microflora.
Gastric or intestinal isografts can be prepared from embryonic day 15 transgenic animals and implanted into the subcutaneous tissues of normal young adult syngeneic hosts, for example, by using previously established methods (Rubin, et al., J. Biol. Chem., 267: 15122-15133 (1992); Rubin, et al., Am. J. Physiol. (Gastroenterol. Liver Physiol., 30), 267, G27-G39, (1994)). These isografts support differentiation of the various gut epithelial cell lineages in the absence of a microflora or pancreatic and biliary secretions. They can be infected with H. pylori by introducing the organism through a cutaneous injection with a sterilized needle and syringe. The effects of the organism in isografts can be assessed in the presence or absence of compounds that may block binding of the bacteria to cellular receptors or alter the course of infection.
The colonization of the intact gut from germ-free and monocontaminated mice, as well as isografts, can be assessed by an in situ binding assay (Falk, et al., Am. J. Physiol. (Gastroenterol. Liver Physiol. 29), 266, G987-G1003), (1994); Falk, et al., Proc. Natl. Acad. Sci. USA 90,2035-2039 (1993)). Furthermore, the degree of mucosal inflammatory response and metaplasia/dysplasia will be evaluated using standard histochemical stains, such as hematoxylin/eosin, periodic acid Schiff (PAS), Alcian blue, and high iron diamine (Baracchini, et al., Histochem. J., 23, 1-9 (1991)). The presence of bacteria on the epithelial cell surface can be detected using standard histochemical methods such as special Giemsa and Warthin-Starry silver stain (Madan, et al., Am. J. Clin. Pathol., 90: 450-453 (1988)).
Additional guidance can be obtained by the following non-limiting examples.
Example 1
.alpha.1,2 FT and .alpha.1,3/4 FT gene sequences
Cloned cDNAs for .alpha.1,2 FT and .alpha.1,3/4 FT were provided in pCDM8 (see, Ausubel, F. M. et al., In Current Protocols in Molecular Biology, Volume 2, pp. 6.13.1-16.13.7 (John Wiley & Sons, New York, 1991)) by Dr. Jan Holgersson (Department of Molecular Biology, Massachusetts General Hospital, Boston, Mass.). These sequences were recovered using polymerase chain reaction (PCR) primers based on the published sequences (Kukowska-Latallo, et al., Genes Develop., 4: 1288-1303 (1990) for .alpha.1,3/4 FT DNA sequence; Larsen et al., Proc. Natl. Acad. Sci. USA, 87: 6674-6678 (1990) for .alpha.1,2 FT DNA sequence; incorporated herein by reference) and human placental DNA as the template DNA. The FT gene sequences were engineered to contain a 5' HindIII site, a Kozak consensus sequence followed by the open reading frame (ORF) for each FT protein of approximately 1.1 kilobases (kb), and a 3' NotI site. The positions of the HindIII and NotI sites are evident in FIGS. 2 and 3. This allowed insertions of the FT sequences into the corresponding HindIII and NotI sites of pCDM8. The PCR cloned .alpha.1,2 FT and .alpha.1,3/4 FT genes have been expressed in COS1 cells and performed the expected transglycosylation reactions.
The cDNA sequences encoding each FT were excised from the recombinant pCDM8 vectors by digestion with HindIII and NotI, and the 1.1 kb HindIII-NotI fragments containing the cloned FT-coding sequences were gel purified using GENECLEAN.TM.(BIO 101 Inc., La Jolla, Calif.). The purified 1.1 kb fragments were then cloned into the HindIII/NotI polylinker sites of pBLUESCRIPT.TM.IISK vectors (Stratagene, La Jolla, Calif.) for easier manipulation. Restriction analysis confirmed that the desired recombinant pBLUESCRIPT.TM. vectors were made (see FIGS. 2 and 3).
The 1.1 kb fragments containing sequences encoding either the .alpha.1,2 FT or the .alpha.1,3/4 FT were released from the recombinant pBLUESCRIPT.TM. vectors by digestion with HindIII and NotI, and the fragments gel purified, as described above, for use in constructing .alpha.1,2 FT and .alpha.1,3/4 FT transgenes.
Example 2
Construction of Transgenes
The 1.1 kb FT HindIII/NotI fragments from the recombinant pBLUESCRIPT.TM. vectors were isolated as described above and blunt-ended with Klenow fragment. The blunt-ended fragments were each gel purified and inserted into the unique BamHI site of vector pLPN.DELTA.B2 which, as shown in FIG. 4, is a derivative of pBLUESCRIPT.TM.I KS and contains nucleotides -596 to +21 of the rat Fabpl promoter sequence (nucleotides 1 to 617 of Sequence ID No. 5), the gene sequence encoding the human growth hormone (hGH) (nucleotides 620 to 2771 of Sequence ID No. 5) and the pgk-neo selection cassette (FIG. 4). Each fragment was inserted into the BamHI site of the pLPN.DELTA.B2 vector, after the opened BamHI site had been blunt-ended with Klenow fragment, thereby forming recombinant plasmids with each fragment at the proper position between the Fabpl.sup.-596 to +21 promoter sequence and the hGH exon 1 sequence. The recombinant plasmids were then transformed into E. coli DH5.alpha. cells. Transformants were screened for recombinant plasmids containing inserts in the proper orientation by digesting plasmid DNAs with AvrII, which generates a 1.6 kb fragment in the case of a correctly oriented insert. Diagrams of recombinant plasmid Fabpl.sup.-596 to +21 /.alpha.1,2 FT/hGH containing the .alpha.1,2 FT sequence in the proper orientation and of recombinant plasmid Fabpl.sup.-596 to +21 /.alpha.1,3/4 FT/hGH containing the .alpha.1,3/4 FT sequence in the proper orientation are shown in FIGS. 5 and 6, respectively.
The integrity of the junctions between the FT fragments and pLPN.DELTA.B2 vector sequences was assessed by sequencing 400 base pairs across the Fabpl.sup.-596 to +21 /FT and FT/hGH junctions, respectively. The sequences at the 5' and 3' junctions were as expected.
Example 3
Microinjection of Embryos
The DNA constructs, Fabpl.sup.-596 to +21 /.alpha.1,2 FT/hGH and Fabpl.sup.-596 to +21 .alpha.1,3/4 FT/hGH, were released from vector DNAs using EcoRI. Four kb fragments containing the -Fabpl.sup.-596 to +21 /.alpha.1,2 FT/hGH or Fabpl.sup.-596 to +21 /.alpha.1,3/4 FT/hGH constructs (see FIGS. 5 and 6) were purified by cleavage of the remaining 4.7 kb vector DNA sequences using Kpnl (for isolating the .alpha.1,2 FT encoding DNA construct) and ScaI (for isolating the .alpha.1,3/4 FT encoding DNA construct), followed by agarose (1%) gel electrophoresis. Each 4 kb fragment was cut out of the gel and gel purified using the GENECLEAN System.TM. (BIO 101, Inc., LaJolla, Calif.). The purity of the Fabpl.sup.596 to +21 /.alpha.1,2 FT/hGH and Fabpl.sup.596 to +21 /.alpha.1,3/4 FT/hGH constructs was confirmed by agarose gel electrophoresis. The concentration of each purified fragment was determined by fluorimetry.
For microinjection, the DNA fragments were diluted to 5 ng DNA/ml in Buffer A(10 mM Tris/0.2 mM EDTA, pH 7.4), and filter sterilized using a 0.22 .mu.m ULTRA-FREE.TM.-MC filter unit (Millipore Corp., Bedford, Mass.). The purified DNA was then injected into the male pronucleus of newly fertilized FVB/N mouse oocytes using standard microinjection methods (see, for example, Hogan et al., Manipulating the mouse embryo, pp. 157-173 (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1986)).
Example 4
Screening for Transgenic Founder Mice Expressing .alpha.1,2 FT and .alpha.1,3/4 FT
The DNA from tail samples of putative transgenic mice has been isolated and probed with .alpha.1,2 FT and .alpha.1,3/4 FT gene sequences to determine by standard DNA blotting methods whether a particular FT gene has been incorporated into the genome of the mice. Alternatively, the Fabpl.sup.-596 to +21 /FT/hGH transgene constructs can be detected in tail DNA by PCR using primers (5'-AGCTGGCCTTTGACACCTACCAGG-3' (nucleotides 1 to 24 of Sequence ID No. 6) (sense strand of exon 1) and 5'-TCTGTTGTGTTTCCTCCCTGTTGG-3' (Sequence ID No. 7) (antisense strand exon 3)) derived from hGH sequences corresponding to intron 2 (a 352 bp fragment), or portions of the coding regions of the human FT DNAs. The PCR reactions for detecting human FT DNAs in transgenic animals employ 5'-CGACAATCACTGACCTATGGCC-3' of Fabpl promoter sequence corresponding to nucleotides 520 to 541 Sequence ID No. 5 (sense strand) and 5'-CGGCGGTCTGGACACAGGATCG-3' (nucleotides 219 to 240 of Sequence ID No. 8), corresponding to nucleotides +116 to +137 of the anti-sense strand of the human .alpha.1,2 FT DNA ORF or open reading frame (a 230 bp fragment), or 5'-GGCCAGGTAGAACTTGTACCGGG-3' (Sequence ID No. 9), corresponding to nucleotides +715 to +737 of the anti-sense strand of human .alpha.1,3/4 FT DNA ORF (an 830 bp fragment). Lines of transgenic animals have been established by crossing founder mice to normal (non-transgenic) FVB/N littermates.
Expression of .alpha.1,2 FT and .alpha.1,3/4 FT in transgenic mice is determined using lectins or MAbs which bind to the specific carbohydrate structures which are predicted to be generated by expression of the FTs. As noted above, a variety of lectins have been identified which can be useful for detecting the presence of various carbohydrate structures found on normal mouse gut epithelial cells (see Falk et al., Am. J. Physiol. (Gastrointest. Liver Physiol. 29), 266, G987-G1003), (1994)). Lectins which are specific for fucosylated glycoconjugates include Ulex europaeus agglutinin I (UEA-1), Anguilla anguilla agglutinin (AAA), and Lotus tetragonolobus agglutinin.
As an alternative to using either lectins or MAbs, H. pylori , itself, is labeled, for example, with fluorescein isothiocyanate or digoxigenin-3-O-succinyl-.epsilon.-aminocaproic acid N-hydroxysuccinimide ester (Dig-NHS, Boehringer Mannheim), and used to detect the newly synthesized surface carbohydrate receptors on the surface of cells of tissue samples of the transgenic mice as described by Falk et al., Proc. Natl. Acad. Sci. USA, 90: 2035-2039 (1993) and Boren et al., Science, 262: 1892-1895 (1993)).
Modifications and variations of the transgenic animal models for screening of antiviral compounds and vaccines, and method of making, will be obvious to those skilled in the art. Such modifications and variations are intended to come within the scope of the appended claims.
Claims
We claim:
1. A transgenic mouse expressing in its gut epithelial cells, under control of a gut epithelial cell specific promoter, an enzyme selected from the group consisting of human GDP-L-fucose: .beta.-D-galactoside 2-.alpha.-L fucosyltransferase and human GDP-L-fucose: .beta.-D-N-acetylglucosamide 3/4-.alpha.-L-fucosyltransferase, wherein Helicobacter pylori adhere to the gut epithelial cells.
2. The transgenic mouse of claim 1 wherein the mouse is gnotobiotic.
3. Gut epithelial cells cultured in vitro from the transgenic mouse of claim 1 or 2, wherein the cells express the enzyme and bind Helicobacter pylori.
4. A method of making a transgenic mouse model or an in vitro mouse gut epithelial cell model for Helicobacter pylori adhesion to gut epithelial cells comprising: introducing into a living mouse embryo DNA molecules comprising the coding sequence for a functional human GDP-L-fucose: .beta.-D-galactoside 2-.alpha.-L-fucosyltransferase or human GDP-L-fucose: .beta.-D-N-acetylglucosamide 3/4-.alpha.-L-fucosyltransferase, operably linked to a gut epithelial cell specific promoter; transferring the embryo to a foster mother; and assaying the resultant born mouse for the adhesion of H. pylori to gut epithelial cells.
5. A method to screen a compound for the ability to inhibit Helicobacter pylori adhesion to gut epithelial cells comprising: administering Helicobacter pylori bacteria to a transgenic mouse of claim 1, or cultured gut epithelial cells derived from the transgenic mouse, wherein the Helicobacter pylori bacteria adhere to gut epithelial cells in the transgenic mouse, or the cultured gut epithelial cells; assaying the gut epithelial cells of the transgenic mouse, or the cultured gut epithelial cells, to determine the number of adherent Helicobacter pylori bacteria on the surface of the cells; administering the compound to be screened to the mouse or cells; and assaying the gut epithelial cells to determine if the compound decreased the number of adherent Helicobacter pylori bacteria.
6. The method of screening of claim 5, wherein the Helicobacter pylori bacteria are administered first to the transgenic mouse, or the cultured cells derived from the transgenic mouse; the Helicobacter pylori are allowed to adhere; and then the compound to be tested is administered to the infected transgenic mouse, or the cultured cells derived from the transgenic mouse.
7. The method of screening of claim 5, further comprising comparing the decrease in number of bacteria adherent to the gut epithelial cells of the transgenic mouse to the decrease in number of bacteria adherent to the gut epithelial cells of a transgenic mouse not exposed to the compound to be screened.
8. The method of screening of claim 7 further comprising exposing the mouse to a stimulus or condition selected from the group consisting of dietary changes, gastro-intestinal pH changes, changes in ambient temperature, changes in body temperature, and chemical induction of an inflammatory response.
9. The method of claim 7 wherein the mouse is vaccinated with a vaccine against Helicobacter pylori about 10 days prior to administration of the bacteria to the mouse.
Non-Patent Literature (42)
- Keller et al (1992) FEBS Letters 313:265-269.
- Strojek et al (1988) Genetic Engineering: Principles and Methods 10: 238.
- Kim et al (1993) Diabetes 42: 1799-1808.
- Kollias et al (1993) In Grosveld et al, eds, Transgenic Animals, p. 84.
- Chandrashekar et al (1994) Mol Androl 6: 95-102.
- Loudon et al (1993) Clin Exp Pharmacol Physiol 20: 283-288.
- RH Whitehead et al (1993) Proc Natl Acad Sci USA 90:587-591.
- GT Merlino (1991) FASEB J: 5:2996-3001.
- Baracchini, et al., "Patterns of Intestinal Metaplasia in Gastric Biopsies. A Comparison of Different Histochemical Classifications", Histochem. J., 23:1-9 (1991).
- Bird, J.M., et al., "Oligosaccharides Containing Fucose Linked .alpha.(1-3) and .alpha.(1-4) to N-Acetylglucosamine Cause Decompaction of Mouse Morulae", Dev. Biol., 104:449-460 (1984).
- Blaser, M.J., "Helicobacter Pylorir: Microbiology of a `Slow` Bacterial Infection", Trends Microbiol., 1:255-260 (Oct. 1993).
- Boren, T., et al., "Attachment of Helicobacter Pylori to Human Gastric Epithelium Mediated by Blood Group Antigens", Science262:1892-1895 (Dec., 1993).
- Clausen, H., et al., "ABH and Related Histo-Blood Group Antigens; Immunochemical Differences in Carrier Isotypes and Their Distribution", Vox Sang., 56:1-20 (1989).
- Evans, D.G. et al., "Cloning, Nucleotide Sequence, and Expression of a Gene Encoding an Adhesin Subunit Protein of Helicobacter Pylori", J. Bacteriol., 175:674-683 (Feb., 1993).
- Falk, P., et al., "Lectins Are Sensitive Tools for Defining the Differentiation Programs of Mouse Gut Epithelial Cell Lineages", Am. J. Physiol. (Gastroenterol. Liver Physiol., 266), G987-G1003, (1994).
- Falk, P., et al., "An In Vitro Adherence Assay Reveals that Helicobacter Pylori Exhibits Cell Lineage-Specific Tropism in the Human Gastric Epithelium", Proc. Natl. Acad. Sci USA, 90:2035-2039 (Mar., 1993).
- Forman, D., et al., "An International Association Between Helicobacter Pylori Infection and Gastric Cancer", (The Eurogast Study Group), Lancet, 341:1359-1362 (May, 1993).
- Gossler, A., et al., "Transgenesis by Means of Blastocyst-Derived Embryonic Stem Cell Lines", Proc. Natl. Acad. Sci. USA, 83:9065-9069 (Dec., 1986).
- Graham, D.Y., et al., "Effect of Treatment of Helicobacter Pylori Infection on the Long-Term Recurrence of Gastric or Duodenal Ulcer", Ann. Int. Med. 116(9):705-708 (May, 1992).
- Granholm, T., et al., "Cytokine Responsiveness in Germfree and Conventional NMRI Mice", Cytokine, 4(6):545-550 (Nov., 1992).
- Hermiston, M.L., et al., "Chimeric-Transgenic Mice Represent a Powerful Tool for Studying How the Proliferation and Differentiation Programs of Intestinal Epithelial Cell Lineages are Regulated", Proc. Natl. Acad. Sci. USA, 90:8866-8870 (Oct. 1993).
- Hogan, B., et al., "Microinjection of DNA Into Pronuclei", In, Manipulating the mouse embryo, 157-173 (Cold Spring Harbor Laboratory , Cold Spring Harbor, NY, 1986).
- Jat, P.S., et al., "Cell Lines Established by a Temperature-Sensitive Simian Virus 40 Large-T-Antigen Gene Are Growth Restricted at the Nonpermissive Temperature", Mol. Cell Biol., 9(4):1672-1681 (Apr., 1989).
- Karlsson, K-A., "Animal Glycosphingolipids as Membrane Attachment Sites for Bacteria", Ann. Rev. Biochem., 58:309-350 (1989).
- Khillan, J.S., et al., "Gene Transactivation Mediated by the TAT Gene of Human Immunodeficiency Virus Transgenic Mice", Nucleic Acids Res., 16(4):1423-1430 (1988).
- Kim, S.H., et al., "Transgenic Mouse Models That Explore the Multistep Hypothesis of Intestinal Neoplasia", J. Cell Biol., 123(4):877-893 (Nov., 1993).
- Kukowska-Latallo, J.F., et al., "A Cloned Human cDNA Determines Expression of a Mouse Stage-Specific Embryonic Antigen and the lewis Blood Group .alpha.(1,3/1,4)Fucosyltransferase", Genes Dev., 4:1288-1303 (1988).
- Levine, T., et al., "The Precancer-Cancer Sequence", In Helicobacter pylori infection (Northfield, Mendall, and Goggin eds.), 88-89 (Kluwer Acad. Publ., Boston (1993).
- Larsen, R.D., et al., "Molecular Cloning, Sequence, and Expression of a Human GDP-L-Fucose:-.beta.-D-Galactoside 2.beta.-L-Fucosyltransferase cDNA That Can Form the H Blood Group Antigen", Proc. Natl. Acad. Sci. USA, 87:6674-6678 (Sep., 1990).
- Madan, E., et al., "Evaluation of Staining Methods for Identifying Campylobacter Pylori", Am. J. Clin. Pathol., 90:450-453 (Apr., 1988).
- Paulson, J.C., et al., "Glycosyltranferases--Structure, Localization, and Control of Cell Type-Specific Glycosylation", J. Biol. Chem., 264(30):17615-17618 (Oct., 1989).
- Roth, K.A., et al., "Regulation of Gene Expression in Gastric Epithelial Cell Populations of Fetal, Neonatal, and Adult Transgenic Mice", Am. J. Physiol. (Gastrointest. Liver Physiol., 26), 263-G186-G197 (1992).
- Rubin, D.C., et al., "Use of Fetal Intestinal Isografts from Normal and Transgenic Mice to Study the Programming of Positional Information Along the Duodenal-to-Colonic Axis", J. Biol. Chem., 267(21):15122-15133 (Jul., 1992).
- Rubin, D.C., et al., "Use of Isografts to Study Proliferation and Differentiation Programs of Mouse Stomach Epithelia", Am. J. Physiol. (Gastroenterol. Liver Physiol., 30), 267:G27-G39, (1994).
- Sakamoto, J., et al., "Expression of Lewis.sup.a, Lewis.sup.b, Lewis.sup.x, Lewis.sup.y, Sialyl-Lewis.sup.a, and Sialyl-Lewis.sup.x Blood Group Antigens in Human Gastric Carcinoma and in Normal Gastric Tissue", Cancer Res., 49:745-752 (Feb., 1989).
- Sanchez, J., et al., "Binding of Bacteria to Carbohydrates Immobilized on Beads to Demonstrate the Presence of Cell-Associated Hemagglutinins in Vibrio Chlorae", Acta Pathol. Microbiol. Immunol. Scand., 98:353-357 (1990).
- Sandgren, E.P., et al., "Overexpression of TGF.alpha. in Transgenic Mice: Induction of Epithelial Hyperplasia, Pancreatic Metaplasia, and Carcinoma of the breast", Cell, 61:1121-1135 (Jun., 1990).
- Sharon, N., et al., "Lectins as Cell Recognition Molecules", Science, 246:227-234 (1989).
- Simon, T.C., et al., "Use of Transgenic Mice to Map cis-Acting Elements in the Liver Fatty Acid-Binding Protein Gene (Fabpl) That Regulate Its Cell Lineage-Specific, Differentiation-Dependent, and Spatial Patterns of Expression in the Gut Epithelium and in the Liver Acinus", J. Biol. Chem., 268(24):18345-18358 (1993).
- Tosh, F.D., "Characterization of a Fucoside-Binding Adhesin of Candida Albicans", Infect. Immun., 60(11):4734-4739 (Nov., 1992).
- Trahair, J.E., et al., "Use of Transgenic Mice to Study the Routing of Secretory Proteins in Intestinal Epithelial Cells: Analysis of Human Growth Hormone Compartmentalization as a Function of Cell Type and Differentiation", J. Cell Biol., 109:3231-3242 (Dec., 1989).
- "Transient Expression of Proteins Using COS Cells", In Current Protocols in Molecular Biology, 2:16.13.1-16.13.7, Ausubel, F.M., et al., eds. (John Wiley & Sons, New York, 1991) by Dr. Jan Holgersson (Department of Molecular Biology, Massachusetts General Hospital, Boston, Massachusetts).